U.S. patent number 5,640,549 [Application Number 08/254,314] was granted by the patent office on 1997-06-17 for method and apparatus for graphical modeling of psychiatric and medical records.
Invention is credited to Seth M. Powsner, Edward Rolf Tufte.
United States Patent |
5,640,549 |
Powsner , et al. |
June 17, 1997 |
Method and apparatus for graphical modeling of psychiatric and
medical records
Abstract
An apparatus and method for determining the course of a
patient's illness and response to treatment. The system uses a
matrix of coordinated, nonlinear time plots to guide diagnosis and
record treatment. The preferred method comprises representing
patient findings and treatments as a matrix of small graphs each
comprised of a label, the numeric value of the most recent plotted
value, plotted values themselves and axis markers. The graphs are
coordinated so that plots in the same column share the same
horizontal time axis and plots in the same row share the same
vertical clinical value axis. Nonlinear scales are used to allow
both recent and historical trends to be presented on the same axis
and to allow various clinical values to be presented on the same
axis. The system also presents a small of amount of textual
information about the patient to identify and annotate the
graphical presentation.
Inventors: |
Powsner; Seth M. (Hamden,
CT), Tufte; Edward Rolf (Cheshire, CT) |
Family
ID: |
22287494 |
Appl.
No.: |
08/254,314 |
Filed: |
June 6, 1994 |
Related U.S. Patent Documents
|
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
Issue Date |
|
|
101981 |
Aug 4, 1993 |
|
|
|
|
Current U.S.
Class: |
703/11 |
Current CPC
Class: |
G06T
11/206 (20130101) |
Current International
Class: |
G06T
11/20 (20060101); G06T 015/00 () |
Field of
Search: |
;364/413.01,413.02,413.05,413.06,578,41M ;128/630,721 ;395/500 |
References Cited
[Referenced By]
U.S. Patent Documents
Other References
S Higgins et al., "A Graphical ICU Workstation" SCAMC 1991
Proceedings. .
Jesse B. Hall, M.D., Gregory A. Schmidt, M.D., & Lawrence D. H.
Wood, M.D., eds, Principles of Critical Care, McGraw-Hill, Inc.,
New York, 1992. The teachings of Chapter 41, "Computer
Applications" by Alan H Morris, M.D. and Reed M. Gardner, Ph.D. at
pp. 500-514. .
Carol L. Lake, M.D., Clinical Monitoring, W. B. Saunders Co.,
Philadelphia, 1990. The teachings of Chapter 12, "Patient Data
Management Systems" by John W. Hoyt, M.D. and Harry Comerchero at
pp. 433-451. .
J. Andrew Billings, M.D. & John D. Stoeckle, M.D., The Clinical
Encounter, Year Book Medical Publishers, Inc., Chicago, 1989. The
teachings of Chapter 6, "Recording" at pp. 81-91. .
Lawrence L. Weed, M.D., Medical Records, Medical Education, and
Patient Care, The Press of Case Western Reserve University,
Cleveland, 1969. The teachings of Chapter 5, "The Progress Notes"
at pp. 49-60 and chapter 6, Flow Sheets at pp. 61-82. .
James D. Foley, Andries van Dam, Steven K. Feiger, and John F
Hughes, Computer Graphics: Principles and Practice, 2nd ed.,
Addison-Wesley Publishing Company, Inc., Reading 1992. The
teachings of Chapter 2, at pp. 25-66, Chapter 3 at pp. 67-144, and
Chapter 7 at pp. 285-346. .
Edward R. Tufte, Envisioning Information, Graphics Press, Cheshire,
CN, 1990. pp. 36-51 and pp. 56-57..
|
Primary Examiner: Teska; Kevin J.
Assistant Examiner: Loppnow; Matthew
Attorney, Agent or Firm: Choate, Hall & Stewart
Parent Case Text
REFERENCE TO RELATED APPLICATIONS
This application is a continuation under 37 C.F.R. .sctn.1.53 of
United States patent application Ser. No. 08/101,981 (Attorney
Docket No. SMD-001), filed Aug. 4, 1993, for "METHOD AND APPARATUS
FOR GRAPHICAL MODELING OF PSYCHIATRIC AND MEDICAL RECORDS", now
abandoned.
Claims
We claim:
1. An apparatus for graphical modeling of at least one of a medical
and a psychiatric history, comprising
A. input means for receiving one or more medical/psychiatric
history signals, each associated with a clinical data point to
modeled, each such history signal having an original temporal value
and an original quantitative value,
B. temporal scaling means for generating a modified temporal value
for each of said medical/psychiatric history signals, each said
modified temporal value being a function of a time-wise location of
the respective original temporal value in an epoch in which that
original temporal value belongs with respect to (n) such epochs,
where (n) is an integer greater than one, each such epoch being
associated with a time period and duration of medical/psychiatric
significance, where both a duration and a time-wise distance from a
time of interest of each epoch (i) decreases with successive
integer values of (i) between one and (n), and
C. gross-quantitative scale means for generating a modified
quantitative value for each of said of medical/psychiatric history
signals, each said modified quantitative value being a function of
the range of values to which the original quantitative value
belongs with respect to (m) designated gross-quantitative ranges,
and
D. modeling means for generating an output signal graphically
representative of the modified quantitative values of said
medical/psychiatric history signals as a function of respective
modified temporal values.
2. An apparatus according to claim 1 wherein said temporal scaling
means includes means for selecting said time of interest to be the
present.
3. An apparatus according to claim 1, wherein said
gross-quantitative scale means includes means for selecting said
(m) designated gross-quantitative ranges to include clinically
normal, abnormally elevated, markedly elevated, abnormally reduced,
and markedly reduced.
4. An apparatus according to claim 1, wherein said temporal scaling
means includes means for selecting said (n) temporal regions to
include epochs representing a current day, a prior week, a prior
month, and a prior period going back to birth.
5. An apparatus according to claim 1, which includes display means
for generating a plot on a computer monitor or printer or plotter
as a function of said output signal.
6. An apparatus according to claim 5, wherein said display means
includes means for normally displaying said (n) temporal regions to
span like distances on said plot.
7. An apparatus according to claim 6, wherein said display means
includes means for displaying at least a selected temporal region
to span greater distances on said plot than others of said temporal
regions.
8. An apparatus according to claim 7, wherein said display means
includes means for selecting said selected temporal regions to be
one representing a prior period of illness.
9. An apparatus according to claim 5, wherein said display means
includes means for displaying portions of said plot in different
respective colors.
10. An apparatus according to claim 5, wherein said display means
includes means for displaying a plurality of said plots in
alignment and with like horizontal axes on said display means.
11. A method for graphical modeling of at least one of a medical
and a psychiatric history, comprising
A. receiving one or more medical/psychiatric history signals, each
associated with a clinical data point to modeled, each such history
signal having an original temporal value and an original
quantitative value,
B. generating a modified temporal value for each of said
medical/psychiatric history signals, each said modified temporal
value being a function of a time-wise location of the respective
original temporal value in an epoch in which that original temporal
value belongs with respect to (n) such epochs, where (n) is an
integer greater than one, each such epoch being associated with a
time period and duration of medical/psychiatric significance, where
both a duration and a time-wise distance from a time of interest of
each epoch (i) decreases with successive integer values of (i)
between one and (n), and
C. generating a modified quantitative value for each of said of
medical/psychiatric history signals, each said modified
quantitative value being a function of the range of values to which
the original quantitative value belongs with respect to (m)
designated gross-quantitative ranges, and
D. generating an output signal graphically representative of the
modified quantitative values of said medical/psychiatric history
signals as a function of respective modified temporal values.
12. A method for graphical modeling of at least one of a medical
and a psychiatric history, comprising
A. receiving one or more medical/psychiatric history data points to
be modeled, each data point being associated with a time value and
a quantitative value,
B. generating a data modelling point having first and second
coordinates, the first coordinate being a function of (i) a
relative location of an epoch in which the time value associated
with that data point belongs with respect to a plurality of epochs,
and (ii) a relative location of that time value within that
epoch,
the second coordinate being a function of a range of values to
which the quantitative value associated with that data value with
respect to a plurality of ranges.
13. A method according to claim 12, wherein step (B) includes
selecting said plurality of ranges to include clinically
significant ranges.
14. An apparatus according to claim 13, wherein step (B) includes
selecting said plurality of ranges to include one or more of
clinically normal, abnormally elevated, markedly elevated,
abnormally reduced, and markedly reduced.
15. An apparatus according to claim 12, wherein step (B) includes
selecting said plurality of epochs to include epochs of clinical
significance.
16. An apparatus according to claim 15, wherein step (B) includes
selecting said plurality of epochs to include one or more of a
current day, a prior week, a prior month, and a prior period going
back to birth.
17. A method according to claim 12, including the step of
displaying said data modelling points.
18. A method according to claim 12, including the step of
displaying said data modelling points on an output device including
any of a computer monitor, a printer, and a plotter.
19. A method according to claim 17, wherein the step of displaying
said data modelling points includes the step of normally displaying
said plural epochs to span like distances as displayed by an output
device.
20. A method according to claim 17, wherein the step of displaying
said data modelling points includes the step of displaying at least
a selected epoch to span a greater distance than others of said
temporal regions as displayed by an output device.
21. A method according to claim 20, including the step of selecting
said selected epoch to be one representing a prior period of
illness.
Description
REFERENCE TO APPENDICES
The disclosure of this patent document contains material which is
subject to copyright and/or mask work protection. The owner thereof
has no objection to facsimile reproduction by anyone of the patent
document or the patent disclosure, as it appears in the U.S. Patent
and Trademark Office patent file or records, but otherwise reserves
all copyright and/or mask work rights whatsoever.
BACKGROUND OF THE INVENTION
This invention relates to an apparatus and methods for determining
the course of a patient's illness and response to treatment. More
particularly, this invention provides a mechanism for analyzing for
presentation in graphical form information concerning the patient's
symptoms, emotional state, physical findings, test results and
treatments. The invention has application in the fields of
medicine, psychiatry, surgery, obstetrics, pediatrics, and other
medical specialties.
In general, written records for medical and for psychiatric
patients have been used to record symptoms as they are reported,
physical findings as they are observed, test results as they are
measured and treatments as they are instituted. Early medical
record systems relied solely on paper documentation. Typically,
medical personnel recorded observations or treatments by hand on
preprinted forms for storage on clipboards, in binders or in
folders. Certain reports, such as lab write-ups, may have been
recorded in typewritten form. More recently, medical records have
been stored in computer databases for on-line access and display in
text format and, to a limited degree, graphical format.
In this regard, reference may be made to the book by Jesse B. Hall,
M.D., Gregory A. Schmidt, M.D., & Lawrence D. H. Wood, M.D.,
eds, Principles of Critical Care, McGraw-Hill, Inc, New York, 1992.
The teachings of Chapter 41, "Computer Applications" by Alan H
Morris, M.D. and Reed M.Gardner, Ph.D. at pages 500-514 are
incorporated herein by reference. Reference may be made to the book
by Carol L. Lake, M.D., Clinical Monitoring, W. B. Saunders Co.,
Philadelphia, 1990. The teachings of Chapter 12, "Patient Data
Management Systems" by John W. Hoyt, M.D. and Harry Comerchero at
pages 433-451 are incorporated herein by reference. Reference may
be made to the book by J. Andrew Billings, M.D. & John D.
Stoeckle, M.D., The Clinical Encounter, Year Book Medical
Publishers, Inc. Chicago, 1989. The teachings of Chapter 6,
"Recording" at pages 81-91 are incorporated herein by reference.
Reference may be made to the book by Lawrence L. Weed, M.D.,
Medical Records, Medical Education, and Patient Care, The Press of
Case Western Reserve University, Cleveland, 1969. The teachings of
Chapter 5, "The Progress Notes" at pages 49-60 and chapter 6, "Flow
Sheets" at pages 61-82 are incorporated herein by reference.
Reference may also be made to the book by James D. Foley, Andries
van Dam, Steven K. Feiger, and John F Hughes, Computer Graphics:
Principles and Practice, 2nd ed., Addison-Wesley Publishing
Company, Inc., Reading 1992. The teachings of Chapter 2, at pages
25-66, Chapter 3 at pages 67-144, and chapter 7 at pages 285-346
are incorporated herein by reference.
In conventional medical and psychiatric record systems, information
is usually presented in one of a small number of forms: prose,
flowsheet, time plot. Hand written prose is the form most often
used to record and present patient information. Sometimes word
processing or automated data processing techniques are used to make
the text more legible and accessible.
Patient information that is largely numeric is sometimes presented
in flowsheets, also referred to as spreadsheets by some. Each
flowsheet column corresponds to succeeding dates or times. Each
flowsheet row is reserved to present a certain type of data such as
temperature or blood pressure or fluids given. Alternatively, rows
may be used for times and columns for types of data.
Sometimes automated data processing techniques are used to make the
values easier to enter and more legible. Conventional time plots
(graphs) are occasionally used in permanent patient records to
present the results of a specific laboratory test over the course
of an illness. The test value is plotted along a linear vertical
scale and time is plotted along a linear horizontal scale. Such
plots are more commonly used for instantaneous display of
electrocardiograms and blood pressure for patients in an intensive
care unit, in an operating room, or in a recovery room. These plots
show a small number of critical values for a short period of time
on a linear scale and are not usually part of the permanent
record.
A problem presented by these prior art systems is that the
presentation of patient information in prose form makes it
difficult to apprehend the course of a patient's disease, or the
course of a patient's treatment. Hand written prose is often
difficult to read. Typing may improve legibility, but it does not
organize the information so that the human eye may perceive a
pattern. Rather, prose presentation requires the eye and the mind
to work continuously to pick and sort out patterns of disease and
treatment from text ordered only by the time it happened to be
recorded.
Further, because a long list of numbers only draws the eye and the
mind along in a straight line, the prior art systems using the
flowsheet form also make it difficult to apprehend the course of
patient's disease, or the course of patient's treatment. Moreover,
the numbers themselves are surrounded by other numbers. The upward
or downward trend, the value to value variability is not directly
perceived.
Presenting patient information in conventional time plots makes it
difficult to apprehend the patient's overall condition.
Conventional time plots do make it possible for the eye to follow
the trends and variations of a given type of information. However,
prior methods only present one or a small number of plots to be
viewed together and linear scaling makes it difficult to perceive
the trends over the last few days in the same glance as the trends
over the years of the patient's life.
An object of this invention, accordingly, is to provide improved
methods and apparatus for analysis and presentation of patient
information which remedies the above mentioned drawbacks. More
particularly, and object of the invention is to provide a mechanism
for analyzing for presentation in graphical form patient
information so that it can be quickly and accurately perceived by
physicians, nurses, and other medical personnel.
A further object of this invention is to provide such a mechanism
which can operate on a wide variety of patient information
including, but not limited to, physical findings, psychiatric
findings, laboratory test results, and various treatments.
Yet another object of this invention is to provide methods and
apparatus for analyzing for presentation in graphical format
patient information that encompasses long periods of time so that
chronic disease conditions and the results of treatments can be
quickly and accurately perceived.
Still another object of this invention is to provide such methods
and apparatus that can serve as components of an on-line patient
information system.
A still further object of this invention is to provide for the
analysis and presentation of patient information in graphical form
and, thereby, to facilitate the exchange of medical information
around the world by minimizing language differences.
SUMMARY OF THE INVENTION
The aforementioned and other objects are attained by the invention,
which provides in one aspect an apparatus for graphical modeling of
medical or psychiatric histories. The apparatus includes an element
for inputting one or more medical/psychiatric clinical data points
to be modeled. Each history signal has a quantitative value
representing, e.g., a laboratory result, and a temporal value
representing, e.g., time of measurement.
A temporal scaling element generates, from the original temporal
value in each history signal, a "modified" temporal value. Each
modified value represents a time-wise location of the respective
original temporal value in an epoch in which that value belongs.
These epochs are associated with time periods of
medical/psychiatric significance, such as, current day, past week,
past month, lifetime prior to last month. Thus, each epoch has a
duration and a time-wise distance from the present which decreases
with successive epochs.
A gross-quantitative scaling element generates, from the original
quantitative value in each history signal, a modified quantitative
value. These modified values indicate the gross-quantitative range
in which the original quantitative value falls, for example,
clinically normal, abnormally elevated, markedly elevated,
abnormally reduced, and markedly reduced.
A modeling element generates an output signal that represents
graphically the modified quantitative values of each history signal
as a function of its respective modified temporal values.
According to a further aspect of the invention, the apparatus can
include a display element that generates a graphical plot based on
the output signal, that is, it plots the modified quantitative
values as a function of their respective modified temporal values.
While the epochs can be plotted to span like distances on said
plot, according to a further aspect of the invention selected
epochs span greater distances for emphasis.
In still another aspect, the apparatus can plot, in vertical
alignment, the output signals associated with plural sets of
history signals.
In still other aspects, the invention provides a method for
operation of an apparatus of the type described above.
These and other aspects of the invention are evident in the
drawings and in the description which follows.
BRIEF DESCRIPTION OF THE DRAWINGS
A further understanding of the invention may be attained by
reference to the attached drawings, in which:
FIG. 1A depicts a computer system for use in connection with a
preferred practice of the invention;
FIG. 1B depicts a preferred medical or psychiatric record charting
system according to the invention;
FIG. 2 depicts the overall data flow within a system embodying the
invention;
FIG. 3 is a flow chart depicting a processing sequence to produce
the output of the presentation sizing processor;
FIG. 4 is a flow chart depicting a processing sequence to produce
the output of the time scaling processor;
FIG. 5 is a flow chart depicting a processing sequence to produce
the output of the finding & treatment scaling processor;
FIG. 6 is a flow chart depicting a processing sequence to produce
the output of the patient data processor; and
FIG. 7 depicts an exemplary medical record model generated by an
apparatus operating in accord with a preferred practice of the
invention.
DETAILED DESCRIPTION OF THE ILLUSTRATED EMBODIMENT
FIG. 1A illustrates a computer system of the type used in
connection with practice of the invention. Computer 106 is
connected to data processing peripheral units comprising a disk
memory 101, a computer communications network 108, a video monitor
107, a keyboard 110, and a pointing device 109. The disk memory 101
serves as a non-volatile storage element for information accessed
by the computer 106. The computer communications network 108 serves
to allow access to non-volatile storage elements and databases or
other peripheral devices among a collection of computers. The
monitor 107, the keyboard 110, and the pointing device 109 provide
an interface between the computer 106 and the user. Specifically,
the monitor 107 presents a graphic display of signals generated by
the computer 106, while the keyboard 110 and pointing device 109
convert typed messages and position signals into computer-readable
form.
The illustrated computer 106 includes functional units comprising
an i/o controller 103, a central processor 104, and a (random
access) memory unit 105. The i/o controller 103 is an interface
between the computer 106 and its peripheral units. The central
processor 104 serves as the primary source of control, arithmetic,
and logic operations within the computer 106. Further, the memory
unit 105 provides volatile, rapid-access storage for the computer
106, particularly for the central processing unit 104.
The illustrated hardware used for practice of the invention can be
selected from any one of many commercially available programmable
digital computers, e.g., preferably, a commercially available
Macintosh II, running the version 6.0.5 operating system, operating
in accord with the teachings below.
FIG. 1B depicts a preferred medical record graphical modeling
system 115 according to the invention. The system 115 includes an
element 126 that controls the obtaining and analysis of data, as
well as the charting of the analyzed data. Input data 123, about an
individual patient, is obtained from the clinical information
system 118. In the course of obtaining that data, the operating
system and graphical support 119 may be activated directly by
controller 126 as well as indirectly by the clinical information
system 118. Those skilled in the art will appreciate that input
data 123 is not necessarily all obtained at once. Rather, as the
apparatus operates to analyze data 124 concerning details of past
patient history, or as the apparatus operates to form the output
chart 122 based on details of past patient data, the control 126
can activate the clinical information system 118 and the operating
system & graphical support 119 as needed to fill-in input data
123.
Input data 123 directly affecting the size and detailed form of the
output chart 122 is kept in the data store 120 once obtained
through the operating system and graphical support 119 or the
clinical information system 118 by the control 126. Like the input
data 123, the general presentation specification, video monitor
data, font data, and printer data 201-204 are not necessarily all
obtained at once; rather that data can be obtained during the
process and used as described below.
FIG. 2 depicts a functional interrelationship between elements of a
preferred embodiment of the invention, as well as the processing of
data by those elements. The general presentation specification, the
video monitor data, the font data, and the printer data, 201-204,
are presented as inputs to the presentation sizing processor 301.
The output of the presentation sizing processor 301, is the active
presentation specification 205. The active presentation
specification serves as input to the time scaling processor 401, as
input to the finding and treatment scaling processor 501, and as
input to the patient data processor 601. The active presentation
specification encodes information about the exact sizes and
placements for the graphical and textual elements of the patient
model, e.g., position, font, and size of the text of the patient's
name, position, and size of the individual graphs in the matrix,
etc.
Those skilled in the art will appreciate that the designated
general presentation specification 201 need not be implemented as
an independent entity. Indeed in a preferred implementation,
specification 201 is defined by way of constants. Likewise, 205 is
preferably implemented as a collection of variables and procedural
information specifying, for example, that the date text is
positioned by subtracting its display width from position of the
room number text.
Referring, again, to the drawing, the time scaling processor 401
uses information about the size of the individual graphs, the fonts
for axis labels, the position of the graphical matrix from the
active presentation specification 205, the date of birth, and times
that findings or treatments were recorded to produce a time axis
scale 206, as discussed in further detail below. The resultant time
axis scale 206 comprises a mapping from the date and time to
display coordinates, a series of labels, and an optional series of
tick marks.
The finding and treatment scaling processor 501 uses information
about the size of the individual graphs, the fonts for axis labels,
and the position of the graphical matrix from the active
presentation specification 205. It also uses information about
finding and treatment scales 207 such as normal value range,
elevated value range, etc. to produce finding and treatment axis
scales 209 for each finding or treatment included in the finding
and treatment display list 208. A finding and treatment axis scale
comprises a mapping from values to display coordinates, a series of
labels, and an optional series of tick marks.
The patient data processor 601, uses the individual patient data
210, to fill in the elements encoded in the active presentation
specification 205. The individual finding and treatment values are
mapped into the graphs within the matrix using the mapping
information encoded in the axis scales 206 and 209. The text log
entries are fit into the column encoded in the active presentation.
The result is an encoding of the individual graphical model 211,
which can easily be output onto the display using the graphical
support functions of the operating system 119 as is customary in
the art.
FIG. 3 is a flow chart depicting a preferred processing sequence to
produce the output of the presentation sizing processor. The
illustrated process begins at initialization phase 302. The extent
of initialization is determined by the particular environment in
which the invention is practiced. By way of example, this step may
entail setting program variables, and particularly, counters and
graphical position pointers and character font tables.
Additionally, input/output processes, e.g., data file accessing
routines may require initialization prior to sizing, etc.
Following initialization, steps 303-307 control the horizontal
dimensions of the display elements. In step 303, the process
accepts signals indicating the type of printer 102 or monitor 107
to be used for display of the finished graphical model. Stored font
data 203 and additional font data available from the operating
system 119 is used to determine the size of the preferred character
font for small text display such as axis labels. The small text
font is used later to label the horizontal axis which is a critical
factor in determining the number of graphs that may be easily
viewed in one row, step 304. Step 304 also takes into account the
relative size of the text log. Step 305 allocates the exact amount
of display space for the text log based on the results of step 304.
Step 306 allocates space for the vertical axis labels taking into
account the shape of the labels and the results of the preceding
steps. Step 307 allocates any remaining display width to avoid
uneven horizontal display element spacing.
With continued reference to FIG. 3, steps 309-314 control the
vertical dimensions of the display elements. Step 309 proceeds in a
fashion similar to step 303 to determine the size of the preferred
character font for large text display such as the patient's name.
Step 310 allocates space for lines of large text font text at the
top. The position of the upper and lower time axes is then
determined in step 311. The display height for the columns of
graphs is then allocated in step 312. Step 313 determines the
number of graphs that may be viewed in one column.
Still referring to FIG. 3, step 315, produces the display size of
an individual graph based on the column and row information
produced in the preceding steps. The critical components of sizing
information produced in steps 303-315 are encoded into the active
presentation specification 205.
FIG. 4 is a flow chart depicting a preferred processing sequence to
produce the output of the time scaling processor. The illustrated
process begins at initialization phase 402. As with step 302, the
extent of initialization is determined by the particular
environment in which the invention is practiced.
The current date is determined in step 403 from data available from
the operating system 119. The end of the current day is normally
used as the upper limit of the time period displayed.
The date of the patient's birth is determined in step 404 from the
clinical information system 118, using appropriate database
functions 117. The patient's birth is normally used as the lower
limit of the time period displayed.
Alternative time periods of interest may be determined by
processing information from the clinical information system.
Depending on the patient's age, history of hospitalizations and
testing, diagnosis, and other clinical considerations various other
periods of interest may be used in the succeeding steps.
With continued reference to FIG. 4, step 405 converts the dates to
julian format in a manner which to allows time of day to be
included along with date. Key times of interest are calculated in
step 406. Key times are those important for labeling the time axis
or mapping to graph coordinates. These are typically the date a
week ago, a month ago, etc. The results of the previous steps are
used in step 407 to produce a mapping, typically a piece-wise
linear mapping, from julian date and time to graph coordinates.
Time axis labels are produced in step 409. The mapping from step
407 and the labels from step 409 along with the individual graph
sizing from the active presentation specification 205, are combined
to form the time axis scale 206.
FIG. 5 is a flow chart depicting a processing sequence to produce
the output of the finding and treatment scaling processor. The
illustrated process begins at initialization phase 502. As with
step 302, the extent of initialization is determined by the
particular environment in which the invention is practiced.
Step 503 takes as an input the finding and treatment display list
208. It uses this list to control the repetition of the remaining
steps 510-519. Step 503 performs one iteration for each of the
findings or treatments to appear in the final display. An
equivalent implementation would repeat the remaining steps for all
the possibilities among the finding and treatment scales 207. The
output of the iterations are the finding and treatment axis scales
209. As with the time axis scale 206, a finding and treatment axis
scale encodes the mapping from values as recorded for a patient to
graph coordinates suitable for display.
With continued reference to FIG. 5, the iterated step 511 retrieves
a finding and treatment scale from among the scales 207. Using
information from the active presentation specification 205, step
512 converts the scale retrieved into display coordinates. Then
step 513 finalizes the mapping to correspond to the position of the
individual graph within the display.
FIG. 6 is a flow chart depicting a processing sequence to produce
the output of the patient data processor 601. The illustrated
process begins at initialization phase 602. As with step 302, the
extent of initialization is determined by the particular
environment in which the invention is practiced, especially with
respect to individual patient data 210 retrieval through the
database functions 117 of the clinical information system 118.
Throughout the steps illustrated in FIG. 6 and described below, it
should be assumed, unless otherwise stated, that the individual
patient data 210 is retrieved through the database functions 117 of
the clinical information system 118.
After initialization, step 603 positions the patient identifying
information and clinical problem text in the graphical model 211
based on the active presentation specification 205. Text font,
size, and style are processed from the active presentation
specification along with the display position itself.
With continued reference to FIG. 6, step 604 positions the vertical
axis labels for each row. Then step 605 positions the horizontal
time axis labels for each column. This puts the display framework
into place for the matrix of small graphs.
Step 606 causes each small graph to be constructed and positioned
in the graphical model 211. It does this by iterating steps 610-619
for each of the findings and treatment in the finding and treatment
display list 208.
Step 607 causes the text log to be constructed and positioned in
the graphical model 211. It does this by iterating steps 620-629
for each of the text log entries. The iteration proceeds from most
recent to earliest (or earliest to most recent) depending upon the
active presentation specification 205. Also based upon the active
presentation specification, a subset of entries are positioned in
the graphical model (the entire set of entries may exceed the
display space allocated). The iteration of steps 620-629 may be
terminated based on text position calculation performed in step
622.
After the processing in step 607 is finished, the graphical model
211 is complete except for the positioning of a control button if
the model is to be used as part of a user interface function 116.
Depending upon the operating system and graphical support 119
available in the particular environment in which the invention is
practiced, the control button may be added to the model or
positioned implicitly by the operation of the graphical support
119.
Steps 610-619 operate under the control of step 606 to produce and
position one small graph for a finding or a treatment according to
the active presentation specification 205, the time axis scale 206,
and the appropriate finding or treatment axis scale 209. Step 611
retrieves the most recently recorded value for the patient finding
or treatment. Step 612 formats and positions the label for the
graph and includes the most recent value for placement in the
graphical model 211. Step 613 positions the axis markers for the
graph in the graphical model based on the active presentation
specification 205 so that the designated points in the time axis
scale 206 and the appropriate finding or treatment axis scale 209
will be visually evident. Step 614 retrieves all the available
patient values and positions markers in the graphical model
according to the mappings of the designated points in the time axis
scale 206 and the appropriate finding or treatment axis scale 209.
In the case of double valued findings, such as the systolic and
diastolic values of a blood pressure, two or more markers may be
positioned. In the case of findings or treatments whose daily
cumulative value is designated to appear in the graphical model by
the finding and treatment scales 207, cumulative values are
computed and line segments or markers appropriately positioned in
the graphical model.
Steps 620-629 operate under the control of step 607 to produce and
position one text entry according to the active presentation
specification 205. Step 621 retrieves and formats the text one
entry. Text formatting includes, but is not limited to, word
wrapping of lines too long for the column display width allocated
and hanging indents to make individual entries visually apparent.
Step 622 positions the formatted text in the graphical model 211.
Step 623 retrieves the signature information associated with the
text entry and positions the signature or initials in the graphical
model. Step 624 retrieves the degrees, M.D. or R.N. or M.S.W.,
etc., associated with the signature and positions the designation
in the graphical model.
FIG. 7 illustrates an exemplary individual graphical model 211
generated by the system described above after the output chart step
122 to a printer 102, via graphical support software and related
operating system calls 119 as customary in the art of graphical
display. Particularly, the model reveals the input data 123
(laboratory results, vital signs, and medications) retrieved from a
clinical information system 118 for a patient presented as a
coherent matrix of plots with labels and text information in a form
readily understood by physicians and their staff.
FIG. 7 specifically illustrates a graphical model output for a
prototypical patient, Mr John Doe, suffering from diabetic
ketoacidosis the week after his children were sick with sore
throats. He was admitted 9.29.92, two days prior to the day the
graphical model was output, 10.1.92. The plots show that his serum
sodium (NA) was decreased, but returning to normal (last value
134), his serum glucose (Glu) was elevated (last value 243), and
that he was treated with two types of insulin (Reg and NPH, last
doses 10 and 15 respectively). Other information also appears on
the plots. Text notes indicate that he was complaining of pain, but
was sleeping better by 10.1.92.
The size and relative proportions of the plots and text in FIG. 7
are governed by the active presentation specification 205 which is
input to the patient data processor 601 as it creates the
individual graphical model 211. The plots are large enough to
convey trends over time, yet small enough so that a number of them
fit on a page with room for a legible titles including the most
recent numeric values. The text at the top is made large enough to
quickly identify the patient, primary problem (diagnosis), date of
output, room number, and other brief clinical information. The text
column on the side presents brief notes in reverse chronological
order so that the most recent note is at the top.
The active presentation specification 205 also governs the size and
placement of the plot axes. The horizontal time axis labels appear
immediately under the two lines of text at the top of FIG. 7 and
are repeated at the bottom of FIG. 7. The actual time axis scale
206 and labels are governed by the output of the time scaling
processor 206. The left most label "-41 y" indicates that based on
the input data 123 the earliest possible information would be from
the patient's birth 41 years before the date of output. "-m"
indicates one month prior. "-wk" indicates one week prior. "10.1"
is the short form of "10.1.92" the day of output. The horizontal
dots on each plot delineate the corresponding time periods.
The vertical axis labels appear on the left of FIG. 7 and are
repeated on the right of the plots, just left of the text column.
The finding and treatment scaling processor 501 outputs finding and
treatment axis scales 209 so that the various laboratory values,
vital signs, and treatments can use the same axis labels. The large
"+" indicates extremely elevated values, the small "+" indicates
moderately elevated values, the unlabeled central region
corresponds to normal values, the small "-" indicates moderately
decreased values, and the large "-" indicates extremely decreased
values. The vertical column of dots on the plots delineate the
vertical regions. The vertical column's horizontal position also
delineates the beginning of the day of output.
The small crosses on the plots indicate individual patient values.
The original temporal value of each finding or treatment is scaled
by the patient data processor 601 along with the original
quantitative value based on the time axis scale 206 and the finding
and treatment axis scales 209 respectively. The scaling determines
the position of the cross for each patient value on the plot.
Treatments with cumulative effects are totaled for each day and a
line segment positioned by the same scaling used for individual
values.
A further understanding of a preferred practice of the invention
may be attained by reference to the software listings provided in
the Appendix hereto. Those listings are in the Common Lisp
programming language making use of the Common Lisp Object System
(CLOS) and the Common Lisp Interface Manager (CLIM) operating under
the Macintosh Operating System version 6.0.5. Those skilled in the
art will appreciate that certain routines referred to in the
Appendix, such as "run," "clim-shared," and "print-u," serve to
initialize the running environment in a manner conventional to the
art.
The foregoing describes a preferred apparatus and method for
graphical modeling of a medical or psychiatric histories achieving
the objects set forth above. Those skilled in the art will, of
course, appreciate that the illustrated embodiment is exemplary
only and that other embodiments incorporating modifications thereto
fall within the scope of the invention, of which ##SPC1##
* * * * *